Degradation of the Fermentation L. casei Factor. II - Journal of the

Brian L. Hutchings, E. L. R. Stokstad, John H. Mowat, James H. Boothe, Coy W. Waller, Robert B. Angier, Joseph Semb, and Y. SubbaRow. J. Am. Chem...
0 downloads 0 Views 473KB Size
10

HUTCHINGS, STOKSTAD, MOWAT,BOOTHE,WALLER,ANGIER,SEMB AND SUBBAROW Vol. 70 [CONTRIBUTION FROM THE

LEDERLE LABORATORIES DIVISION, AMERICAN CYANAMID

COMPANY]

Degradation of the Fermentation L. casci Factor. I1 BY BRIANL. HUTCHINGS, E. L. R. STOKSTAD, JOHN H. MOWAT, JAMES H. BOOTHE,COY W. WALLER, ROBERTB. ANGIER,JOSEPH SEMB AND Y. SUBBAROW I n the previous paper' data were presented indiBy appropriate manipulations the aromatic cating that 2-amino-4-hydroxypteridine-6-car-amine fragment could be purified. The aromatic boxylic acid, 2-amino-4-hydroxy-6-methylpteri-amine nitrogen as determined by the method of dine and p-aminobenzoic acid were degradation Bratton and Marshall was found to be approxiproducts of the fermentation L. casei factor. I n mately 25% of the total nitrogen. The remaining this paper further degradation products are pre- 7570 of the nitrogen could be converted into asented and evidence for the structure of the liver amino acid nitrogen by acid or alkaline hydrolysis. L. casei factor is outlined. From such hydrolysates the aromatic amine was When the fermentation L. casei factor was dis- isolated and identified as p-aminobenzoic acid. solved in water a t PH 4.0and autoclaved a t 120' Microbiological assay of the hydrolysate indicated for six hours, the activity of the compound was the presence of 3 moles of glutamic acid. Anadestroyed. From this inactive solution a crystal- lytical figures suggested that the amino acids are line product was isolated. The analysis, chemi- joined in peptide linkage as p-aminobenzoyldiglucal properties, and melting point suggested that tamylglutamic acid. The mode of linkage of the the substance was I-pyrrolidonecarboxylic acid. glutamic acids will be discussed in a subsequent However, a cryoscopic determination of the mo- communication. lecular weight offered the possibility of a diketoAfter sulfurous acid hydrolysis of the fermentapiperazine of glutamic acid. tion L. casei factor no one or two carbon-containSynthetic I-pyrrolidonecarboxylic acid caused ing fragments could be detected. Aeiobic ultrano depression of the melting point of the unknown violet inactivation of the fermentation L. casei compound. Further corroborative evidence of factor which cleaves the compound into a pterithe identity of the degradation product with 1- dine fraction and an aromatic amine fraction pyrrolidonecarboxylic acid was furnished by com- failed to yield any carbon dioxide. The evidence parison of their infrared absorption spectra (Fig. 1). suggests that the carbon content of the fermentaWhen the fermentation L.casei factor was dis- tion L. casei factor is that represented by the solved in water and the resulting solution made degradation products. The fermentation L. casei factor was not meas0.5 N with sulfur dioxide, the growth-promoting properties of the compound were destroyed. urably inactivated by hydrogenation over regular There was a marked increase in the fluorescence of Raney nickel, Raney nickel containing aluminum, the solution and an aromatic amine was formed or over palladium on barium sulfate if the hydrothat could be detected by the method of Bratton genation was carried out at a p H of 7.0 or greater and Marshall.2 and at atmospheric pressure. If the compound After removal of the sulfur dioxide, the fluores- was hydrogenated a t p H 1.0 or 3.0 and a t atmoscent moiety reacted rapidly with such typical pheric pressure using a palladium-barium sulfate aldehyde reagents as hydroxylamine, phenyl- catalyst, there was rapid biological inactivation hydrazine and semicarbazide t o form insoluble with the simultaneous formation of an aromatic products. The condensation products exhibited amine. The ultraviolet absorption a t 365 mp was no fluorescence. These derivatives were ex- greatly decreased indicating the reduction of the tremely insoluble and quite unstable but analyses pteridine to a compound that exhibited no abindicated the condensation product with phenyl- sorption a t this wave length. The absorption hydrazine to be a hydrazone. If the fluorescent and fluorescence of the pteridine moiety could be compound arising from sulfurous acid hydrolysis regenerated by oxidation with manganese dioxide was allowed to stand in dilute alkali anaerobically, in acid solution. approximately equal amounts of 2-amino-4The reductive reaction can be carried out chemihydroxypteridine-6-orboxylicacid and a com- cally with zinc dust and either dilute hydrochloric pound which was identified as 2-amino-4-hydroxy- or sulfuric acids, From chemically reduced solu6-methylpteridine were formed. The prcsence of tions of the fermentation L. casei factor, %amino2-amino-4-hydroxy-6-methylpteridinewas unexpected and its mode of formation is obscure. 4-hydroxy-6-methylpteridinehas been isolated The ability to form insoluble precipitates with and characterized. I n a summation of the data presented in this typical aldehyde reagents and the dismutation in dilute alkali indicate the presence of an aldehyde and the previous paper of this series the following group in the pteridine arising from sulfite cleavage salient points serve as a guide to the formulation of the structure of the liver L. cas& factor: of the fermentation L. casei factor. A. Aerobic alkaline hydrolysis, sulfite cleav(1) Stokstad, et J.,THIS JOURNAL, 70, 5 (1948). age and chemical or catalytic reduction each give (2) Bratton and Marshall, J . B i d . Chcm., 128, 537 (1930).

DEGRADATION OF FERMENTATION Lactobacillus casei FACTOR

Jan., 1948

3600

3200

Fig. 1.-Infrared

2000 1800 + 1600 1400 1200 1000 Frequency, crn.-l. absorption spectra of l-pyrrolidonecarboxylic acid: A, synthetic; B, natural.

2800

2400

11

800

rise t o a pteridine fraction and a primary aromatic was evaporated and the residue sublimed in high vacuum 148' for six hours. The sublimate was contaminated amine. This indicates that the point of linkage at with a brown pigment that was removed by extracting the is probably through the aromatic amine group to sublimate with several 0.1-ml. portions of ice-cold ethanol. the pteridine. The residue was re-sublimed at 138' for six hours. There B. A single carbon atom serves as a linkage was obtained a colorless crystalline sublimate (wt. 28 mg.) melted a t 147-148'. A cryoscopic determination between the pteridine and aromatic amine. This which of the molecular weight using camphor as the solvent gave is indicated by obtaining either 2-amino-4-hy- a value of 284. droxypteridine-6-carboxylic acid or 2-amino-4Anal. Found: C, 47.05; H, 5.14; N, 10.93. hydroxy-6-methylpteridine or a compound that Titration data for the compound indicated the presence could be converted into them as degradation prod- of a strong carboxylic acid group. The equivalent weight ucts in several methods of cleavage. Further in- calculated at pH 7.0 was 129.2. The compound exno ultraviolet absorption. The substance gave no dicative evidence is the inability to detect any one hibited color with ferric chloride in aqueous or alcoholic solution, or two carbon-containing fragments arising from The sample adsorbed no hydrogen over Adams platinum two degradative reactions. The evidence sug- oxide catalyst. The compound contained no a-amino acid nitrogen. gests that the one carbon is present as a methylene However, when 1.794 mg. of the compound was dissolved group. Thus, the necessity of oxygen for the in 2 ml. of 1 N sodium hydroxide and heated at 100" for alkaline cleavage is more rational. The reduc- three hours, the following figures were obtained : total tive formation of 2-amino-4-hydroxy-6-methyl-nitrogen, 0.196 mg.; a-amino acid nitrogen before hydrolysis, negative; a-amino acid nitrogen after hydrolypteridine is in accord with this hypothesis. 0.174 mg. Because of the small sample used and the C. The aromatic amine arising from sulfite- sis, accuracy of the Van Slyke determination, these values are cleavage is a tetrapeptide composed of one mole of indicative of complete conversion of the nitrogen into a p-aminobenzoic acid and three moles of glutamic amino acid nitrogen. A semi-quantitative microbiological determination of acid. In the previous paper of this series i t was glutamic acid content of a 1 ili sodium hydroxide observed that in degrading the fermentation L. the hydrolysate of the sublimate was carried out using Lactocusei factor to the dl-liver L.casei factor, two moles bacillus casei as the test organism. A glutamic acid control of a-amino acid nitrogen were liberated. Thus, was run under identical conditions to determine the extent the liver L. cusei factor contains one mole of glu- of racemization. This factor was used in calculating the The sublimate analyzed for 81.5% glutamic acid. tamic acid and the fermentation compound three results. I-Pyrrolidonecarboxylic acid was synthesized according moles of glutamic acid. to the procedure of Abderhalden and Kautzsch.6 The 1Aqueous Hydrolysis of the Fermentation L . casei Factor.-Two hundred mg. of the fermentation L.casei factor was dissolved in 125 ml. of water and autoclaved at 120" for six hours. The resulting clear solution was allowed to stand a t 5" for twenty-four hours. The material that precipitated was centrifuged out and discarded. The water solution was concentrated to dryness and the residue extracted with 4-15 ml. portions of hot absolute ethanol. The ethanol extracts were concentrated to dryness and the residue extracted with 125 ml. of acetone. The acetone

pyrrolidonecarboxylic acid melted not sharply at 141148O.' A mixture of t-pyrrolidonecarboxylic acid and the unknown melted at 115-147'. A cryoscopic determination of the molecular weight c,f the synthetic l-pyrrolidonecarboxylic acid gave a value of 284 indicating polymerization of the compound. Reaction of the Sulfite-cleaved Pteridine with Phenylhydrazine.-Otic hundred tng. of thc fermentation I,. casei factor was dissolved in 50 ml. of water with slight warming and a solution of sulfurous acid added. Thc final volume was 200 ml. and was 0.5 N with rcspect to sulfurous acid. The solution was kept a t 32" for sixteen hours. The solu-

(3) All concentrations carried out iu vacuum. ( 4 ) All melting points are uncorrected. (Z) The ultraviolet absorption was determined with a Beckman Spectrophotometer. T h e fluorescence of the pteridines was determined in 0.05 Ab sodium borate buffer with a Pfaltz and Bauer Fluorophotometer.

(6) Abderhalden and Kautzsch, Z . p h r s i o l . Chem., 68, 487 (1910). (7) On further fractional cryrtallizalion t h e m. p. of the synthetic A lack of niaterinl precluded compound was raised t o l(iO-lG2'. further purification of the isolated I-pyrrolidonecarboxSlic acid. Partial racemization could account for the l o w m . p . of the istdatcd compound (see ref. 3).

Experirnental3.4J

HUTCHINGS, STOKSTAD, MOWAT, BOOTHE, WALLER, ANGER,SEMB AND SUBBAROW Vol. 70

12

tion was concentrated to one-half volume under nitrogen t o remove the sulfur dioxide, warmed to 80" and a solution of phenylhydrazine in dilute acetic acid was added. There was an immediate formation of a red precipitate. The solution was warmed for thirty minutes, then cooled and centrifuged. The precipitate was dissolved in a mixture of piperidine and water (0.4:lO) and precipitated by the addition of acetic acid. The solution and precipitation were repeated twice. Anal. Calcd. for CISHl1N~O:C, 55.5; H, 3.92. Calcd. for CmHlrNeO: C, 48.1; H, 3.41. Fougd: C, 54.2; H,3.53. The analyses indicate the formation of a hydrazone rather than a dihydrazone. Effect of Alkali on the Sulfite-cleaved Pteridine.-Two hundred and fifty mg. of the fermentation L. cas& factor was treated with sulfurous acid as described above. After hydrolysis the solution was concentrated to dryness under nitrogen and the residue dissolved in 6.0 ml. of 1 N sodium hydroxide. The flask was evacuated and then stored in a vacuum desiccator at 4". After fourteen days the precipitate which had formed was centrifuged off and washed with 2 N sodium hydroxide. The absorption spectra of this precipitate in both 0.1 N sodium hydroxide and 0.1 N hydrochloric acid were identical with the absorption spectra of the 2-amino-4-hydroxypteridine-6-carboxylicacid. The amount of the 2-amino-4-hydroxypteridine-6carboxylic acid as determined by ultraviolet absorption was 17mg. The solution after removal of the precipitated 2-ammo4-hydroxypteridine-6-carboxylicacid was adjusted to @H 7.0 and a volume of 30 ml. and extracted five times with

24

ten volume portions of butanol. The total amount of pteridine extracted by the butanol as measured by fluorescence was 30 mg. The butanol was concentrated and'the compound precipitated by the addition of ether. The precipitate was dissolved in 20 ml. of hot water at PH 7.0. A small amount of brown insoluble pigment was removed by centrifugation. The solution was cooled and the precipitate collected; yield was 25 mg. The pteridine was crystallized from 2.0 ml. of 5 N sodium hydroxide. The crystalline precipitate was collected and recrystallized from 1.3 ml. of 5 N sodium hydroxide. The crystals were suspended in water, adjusted to H 7.0 and dissolved by heating. After cooling the pteridne was collected. An4Z. Calcd. for ClHlNsOz: C, 43.5; H, 3.63; N, 36.3. Calcd. for CIHINsO: C, 47.4; H, 3.95; N, 39.6. Found: C,47.4; H,4.43; N,39.6. A comparison of the ultraviolet absorption spectra of this compouna with the absorption spectra of various pteridines indicated its identity with 2-amino-4-hydroxy-6methylpteridine (Fig. 2). The synthesis of this compound will be described in the following paper of this series. The infrared absorption spectra of the natural and synthetic 2amino-4-hydroxy-6-methylpteridineare identical (Fig. 3). The pteridine remaining in the water layer after butan91 extraction was precipitated twice by silver at @H 5.0 to remove it from the amine in solution. The yield was 10 mg. as determined by ultraviolet absorption. The absorption spectra of this pteridine were similar to those for the 2amino-4-hydroxypteridine-6-carboxylic acid. The total amount of pteridines obtained from the fermentation L. casei factor that had been cleaved with sulfurous acid and subsequently treated anaerobically with 1 N sodium hydroxide is presented in Table I.

TABLE I PTERIDINE OBTAINEDON ALKALINE DISMUTATION

18

Milligrams

"

2 x l2

Fermentation L. cas& factor used Amine (calcd. as PABA)

250

48 2-Amino-4-hydroxypteridine-6-carboxylic acid a. Precipitated by 1.0 N sodium

Y

6

b.

0 220

260

340

300

380

420

A, mw Fig. 2.-Ultraviolet absorption spectra of 2-amino-4hydroxy-6-methylpteridine: , in 0.1 N sodium hy, in 0.1 N hydrochloric acid. droxide;

----

-

hydroxide Remaining after extraction with butyl alcohol and precipitated by silver

Moles per mole of

F. L. C. F.

0.98

17

-23

10 -

.14 .37

Total 27 2-Amino-4-hydroxy6-methylpteridine a. Total amount extracted by butanol 30

.47

2OOO 1800 1600 1400 1200 loo0 800 Frequency, cm.-*. Fig. X--Infrarcd absorption spcctra of 2-amino-4-hydroxy-6-methylpteridine: A, natural; B. synthetic. 3600

3200

2800

2400

Jan., 1948

DEGRADATION OF FERMENTATION hCi?ObUC&4S

Purification of the Aromatic Amine.-One gram of the fermentation L. cask factor was hydrolyzed with sulfurous acid. After cleavage, the solution was concentrated to one-half volume to remove the sulfur dioxide. The pteridine fraction was removed by precipitation with silver sulfate at pH 2.0. Excess silver and sulfate ions were removed from the filtrate; yield was 115mg. calculated as p-aminobenzoic acid. The solution was adjusted to pH 2.0 and concentrated to dryness. The residue was dried thoroughly and the amine extracted from the solid with absolute isopropanol. The isopropanol solution was cbncentrated to dryness and the residue dissolved in 30 ml. of water. Excess lead acetate was added. After standing overnight at 5 ' , the lead precipitate was collected and washed with 5 ml. of cold water containing a little lead acetate. The precipitate was suspended in water and the lead removed as the sulfide; yield was 96.5 mg. calculated as p-aminobenzoic acid. The solution was adjusted to pH 2.0 and concentrated to dryness. The amine was extracted into absolute isopropanol. The isopropanol was evaporated and the residue dissolved in 50 ml. of water. Barium hydroxide was added to fiH 10.0 and the slight precipitate removed. Ethanol was added to 50% concen' overnight. The batration and the material kept at 5 rium precipitate was collected.,dissolved in water, adjusted to pH 2.0 and concentrated to dryness. The amine was extracted with isopropanol and then precipitated by the addition of one volume of petroleum ether. The precipitate was centrifuged off, dissolved in 35 ml. of water and concentrated slightly to remove the isopropanol and petroleum ether. Barium hydroxide was added to pH 10.0 and ethanol to 50% concentration. After thorough chilling the precipitate was collected, washed thoroughly with 50% ethanol, absolute ethanol and acetone; yield was 71 rng. calculated as 9-aminobenzoic acid. Anal. Calcd. for C22HlaN,Oll: C, 50.5; H, 5.35; ND 10.69. Found: C,29.42;H,4.20; N,6.64; ash,50.00Cor. for C in ash: C, 32.46; H, 4.20; N, 6.64; Bas 34.80. Ba-free: C, 49.8; H, 6.44; N, 10.17. As the ash is barium carbonate, the carbon values have been corrected for theamount of carbon retained in the ash. The free acid is extremely soluble in water or alcohol. The compound could not be crystallized from various solvents or combinations thereof. Aromatic Amine Nitrogen.-Duplicate weighings were dissolved equivalent to 13.62micrograms of total nitrogen per ml. ; 29.1 microgram equivalents of p-aminobenzoic acid was found per ml., which is equivalent to 2.97 micrograms of aromatic amine nitrogen. Therefore 22.39% of the total nitrogen was present as aromatic amine nitrogen. Effect of Acid and Alkaline Hydrolysis on the Amine Fragment .-All samples were hydrolyzed for sixteen hours at 120' in a sealed tube. One normal sodium hydroxide, 5 N sodium hydroxide, 1 N hydrochloric acid and 2.5 N hydrochloric acid liberated 57.6, 73.0, 72.2, and 74%, respectively, of the nitrogen as a-amino acid nitrogen. Glutamic acid analysis of the 2.5 N hydrochloric acid hydrolysate was carried out using Lactobacillus arabinosus. Anal. Calcd. for 3 moles of glutamic acid in the amine fragment (as the free acid): 84.25. Found: 81.8,82.5. Isolation of p-Aminobenzoic Acid from the N, Amine.One hundred and eighty-seven mg. of the amine fragment was dissolved in 6 ml. o€5 N sodium hydroxide and hydrolyzed anaerobically for thirteen hours a t 120'. The resulting hydrolysate was diluted to 100 ml. and adjusted to pH 3.0. The solution was extracted with 100 ml. of ethyl acetate and the ethyl acetate extract washed with 25 ml. of water. The ethyl acetate was removed by distillation and the residue dissolved in 0.5 ml. of hot water. On cooling the compound crystallized as needles. The pre-

ulS&

FACTOR

13

cipitate was recrystallized from 0.25 ml. of hot water, m. p. 180.5-182.5". I n admixture with an authentic s p e c h e n -aminobenzoic acid (m. p. 183.5-184.5') it melted at Ok! 1

5-183.5'. Anal. Calcd. for CrHfiN: C, 61.3; H, 6.11. Found: C, 61.5; H, 4.93. Detection of One or Two Carbon-atom-cOntaining Fragments.-From sulfurous acid hydrolysates of the fermentation L. casei factor no carbon dioxide, glyoxal, glycolic aldehyde or ethylene glycol could be detected by appropriately controlled tests. No carbon dioxide was formed on aerobic ultraviolet irradiation of the fermentation L. casei factor. Reductive Cleavage of the Fermentation L. cask Fac-

tor.-Two hundred and sixteen mg. of the fermentation L. casei factor was dissolved in 250 ml. of 0.4 N sulfuric acid and 5 g. of zinc dust was added. After twenty-five minutes the excess zinc dust was removed by filtration and 5 g. of manganese dioxide added. After ten minutes of oxidation the solution was filtered and enough dipotassium hydrogen phosphate was added to the filtrate to raise the p H to 6.5. The zinc phosphate was centrifuged off, washed with water and discarded. The centrifugate and washings were combined The distribution of the pteridine between butanol and water at various pH values was determined. The results suggested that the compound contained no carboxylic acid group. The solution (pH 6.5) was extracted with three ten-volume portions of butanol. The butanol extracts were concentrated and the pteridine precipitated by the addition of ether. The compound was crystallized twice from 4 N sodium hydroxide, converted to the free compound and dried. The ultraviolet absorption spectra of the isolated pteridine were identical with the spectra of an authentic sample pf the 2-amino-4-hydroxy-6-methylpteridine.

Acknowledgment.-It is a pleasure to acknowledge the assistance of the Misses E. Boggiano, B. Eames, E. Brogen and A. Bux6 for the microbiological assays and the spectrophotometric determinations. The authors extend their appreciation to Dr. R. C. Gore of the Stamford Research Laboratories, American Cyanamid Company, for the infrared determinations, and to Mr. L. Brancone and co-workers for the microanalyses. The authors are especially indebted to Dr. J. H. Williams for his constant interest and counsel and for his efforts in co6rdinating the work performed in the various laboratories.

summary I-Pyrrolidonecarboxylic acid, 2-amino-4-hydroxypteridine-6-carboxylic acid, 2-amino-4-hydroxy-6-methylpteridineand p-aminobenzoic acid have been isolated as degradation products of the fermentation L. casei factor. Evidence for the existence of 2-amino-4-hydroxypteridine-6-carboxaldehyde and glutamic acid has been presented. The significance of these products in formulating the structure of the L. casei factors has been discussed. PEARL FUWR,NEW YORK RECEIVED JANUARY 24,1947